JPH09146622A - Method and device for generating track of robot - Google Patents

Abstract

PROBLEM TO BE SOLVED: To transform a bent track of the tip of a robot arm on a teaching point into a smoothly curved track and to generate the transformed track. SOLUTION: A teaching track (x) passing teaching points P1 to P3 is transformed into a smooth continuous track x0 passing the teaching point P2 by continuation processing based upon a specified continuous function. Then the track x0 is transformed into a corrected continuous track xi which passed smoothly the inside of the teaching point P2 by continuation correcting processing based upon a correcting continuous function. In the continuation processing, inner division points between the time change waveforms q1 , q2 of respective axes of the robot and extended time change waveforms qn1 , qn2 extended from the contact of the waveforms q1 , q2 are found by a smooth S-shaped continuation function between continuation start time Ts and continuation end time Te which are set up at an equal time interval from the teaching point P2 to obtain the continuation time change waveforms (q) of respective axes and a robot track corresponding to the operation of the axes is set up as the continuation track.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a trajectory generating method and a trajectory generating apparatus for a robot that interpolates between input teaching points to generate a target trajectory.

[0002]

2. Description of the Related Art In recent years, robots have been used in various work sites, but smooth operation and high speed are important.

A conventional robot trajectory generation device will be described below. In the trajectory that the robot passes through to perform the specified work, a plurality of teaching points to be passed are registered in advance by the operator, and the passing points between the teaching points are interpolated by straight lines, arcs, parabolas, etc. It is decided by. The teaching point data giving the information of the teaching point specifies the position and the attitude coordinate of the teaching point, the speed of passing between the teaching points, and the like.

In the case of simply making a linear motion from one teaching point to the next teaching point, a linear trajectory connecting the teaching points is obtained by interpolation calculation between the teaching points. Further, when moving linearly between the three teaching points, when the bending angle of the polygonal line is large, the speed of the tip portion at the node of the polygonal line greatly changes due to the acceleration due to the direction change, which causes the occurrence of tip vibration. . In addition, since a large acceleration acts on the joint axis of the robot when passing through the break point, it shortens the service life of the robot components such as the reducer or causes damage.

In order to avoid such a problem, means for temporarily stopping the robot at the break point of the broken line, and Japanese Patent Laid-Open No.
As disclosed in Japanese Unexamined Patent Publication No. 27443 and Japanese Unexamined Patent Application Publication No. 4-111006, a track that does not pass through a break point of a broken line but connects two line segments that form the broken line with a smooth curve and goes around a node There is a means for generating.

[0006]

In such a conventional robot trajectory generating apparatus, the means for temporarily stopping at the break point of the polygonal line is about twice as long as the case of continuously operating at a constant speed. Need operating time.

Further, in the means for operating on the path that goes around the break point of the broken line, the moving speed is determined for each interpolation section sandwiched between the teach points, so that when the specified speed is different before and after the teach point. Causes discontinuous change in speed, which causes vibration.

The present invention solves the above problems, and an object of the present invention is to provide a trajectory generating method and a trajectory generating apparatus for a robot that can smoothly change the angle of each axis at a teaching point and shorten the operation time. And

[0009]

According to a first aspect of the present invention, there is provided a teaching point which gives a reference of a trajectory along which a tip of a robot moves, a movement command for designating a trajectory shape based on the teaching point, and the trajectory. On a teaching trajectory specified by teaching point data, which includes a moving speed that specifies the speed at which the tip moves above, and an allowable path error that specifies the maximum value of the error between the teaching point and the trajectory. Continuous first teaching point, second
With respect to the teaching point and the third teaching point, by interpolation calculation according to the trajectory shape and the moving speed designated by the teaching point data, each robot axis corresponding to the movement from the first teaching point to the third teaching point The time change waveform of the angle and the moving time are obtained, and based on the moving time, a passing time point called a continuation start time point before the second teaching point and a continuation end time point after the second teaching point are respectively set from the second teaching point. It is set at a time point separated by time, and the first time-varying waveform of each axis of the robot corresponding to the movement from the first teaching point to the second teaching point is in contact with the second teaching point at the second teaching point and linearized until the continuation end point. A first extension time change waveform given by a line segment that changes to
A second extension given by a line segment that touches the second time-varying waveform of each axis of the robot corresponding to the movement from the teaching point to the third teaching point at the second teaching point and linearly changes until the continuation start time point. A time-varying waveform is obtained, and an internal division point between the first time-varying waveform and the second extension time-varying waveform and an internal division point between the second time-varying waveform and the first extension time-varying waveform are predetermined. Is calculated by the internal division ratio of the continuous function of S, and the continuous function is an S-shaped smooth function that gives an internal ratio of 0 at the start of the continuation and 1 at the end of the continuation. As a result, the movement trajectory of the robot according to the angle of each robot axis given by the internal division point from the continuation start time to the continuation end time passes smoothly through the second teaching point and the teaching trajectory. To be a continuous orbit connected to It is a trajectory generation method of bot.

As a result, the continuous trajectory generated is the second
It becomes a curve that smoothly passes the teaching point. Claim 2
The present invention relates to a robot which adds a continuation correction amount based on a slope change rate of a first time change waveform and a second time change waveform at a second teaching point and a predetermined continuation correction function to an internally dividing point. The value of the correction continuous time change waveform of each axis, the robot trajectory corresponding to the correction continuous time change waveform of each axis,
The trajectory generation method for a robot according to claim 1, wherein a corrected continuous trajectory is provided that goes around the second teaching point and is smoothly connected to the teaching trajectory.

As a result, the generated corrected continuous trajectory becomes a curve that smoothly passes inside the second teaching point. In the present invention according to claim 3, a smooth function having 0 at the continuation start time point and the continuation end time point and having the maximum value at the time point when the second teaching point has passed is defined as the continuation correction function, and The trajectory generation method for a robot according to claim 2, wherein a value weighted with a gradient change rate of the first time-varying waveform and the second time-varying waveform is used as the continuation correction amount.

This makes it possible to correct the inner circumference corresponding to the magnitude of the change in the teaching trajectory at the second teaching point. Further, in the invention according to claim 4, the continuation correction amount at the angle of each axis of the robot corresponding to the set continuation disclosure time point and the continuation end time point is obtained in the vicinity of the second teaching point, and the continuation correction amount is determined. The maximum path error between the teaching path and the corrected continuous path is calculated based on the above, and when the maximum path error is equal to or more than the allowable path error given by the teaching point data, the continuation start time point and the continuation end time point are reset. Then, the trajectory generation method for a robot according to any one of claims 2 to 3, wherein the maximum path error does not exceed the allowable path error.

[0013] With this, it is possible to make continuity and correction continuity so that the maximum path error does not exceed the allowable path error.
In the present invention according to claim 5, the continuation time is reset before the arithmetic processing for determining the continuation time change waveform and the corrected continuation time waveform after setting the continuation disclosure time point and the continuation end time point. The robot trajectory generation method according to claim 4 is as described above.

Thus, it is possible to avoid useless continuous processing and correction continuous processing. According to a sixth aspect of the present invention, a teaching point that gives a reference of a trajectory along which the tip of the robot moves, a movement command that specifies a trajectory shape based on the teaching point, and the tip moves on the trajectory. The first continuous trajectory on the teaching trajectory designated by the teaching point data having a moving velocity designating the velocity to be controlled and an allowable path error designating the maximum value of the error between the teaching point and the trajectory.
For the teaching point, the second teaching point, and the third teaching point, the movement from the first teaching point to the third teaching point is supported by interpolation calculation according to the trajectory shape and the moving speed designated by the teaching point data. Interpolation calculation means for calculating a time-varying waveform of the angle of each axis of the robot, and a start point of continuation before and after the second teaching point, based on a moving time from the first teaching point to the third teaching point, Continuing time setting means for setting a passing time called a continuation ending time at a time point equidistant from the second teaching point, and the robot corresponding to the movement from the first teaching point to the second teaching point. A first extension time change waveform given by a line segment that is in contact with the first time change waveform of each axis at the second teaching point and changes linearly until the continuation end time point, and the second teaching point to the third teaching point The robot corresponding to the movement to Obtains a second extended time-varying waveform given by a line segment that varies linearly to the continuous reduction start time in contact with the second teaching point to the second time change waveform of each axis, the first
An internal division point between the time-varying waveform and the second extension time-varying waveform and an internal division point between the second time-varying waveform and the first extension time-varying waveform are obtained by an internal division ratio by a predetermined continuation function. And a time-varying waveform continuation means for providing a continuous time-varying waveform of the angle of each axis of the robot, wherein the continuation function has an internal division ratio of 0 at the start of the continuation and an internal division ratio at the end of the continuation. Is a smooth S-shaped function, the movement trajectory of the robot according to the angle of each axis of the robot given by the internal division point from the continuation start time to the continuation end time is defined as A trajectory generating apparatus for a robot, which is a continuous trajectory that smoothly passes through a teaching point and is smoothly connected to the teaching trajectory.

Thus, it is possible to realize a trajectory generating device for a robot which generates a curved trajectory smoothly passing through the second teaching point. The present invention according to claim 7 provides a rate of change between the first temporal change waveform and the second temporal change at the second teaching point,
The correction selecting means is provided with a continuation correcting means for obtaining a continuation correction amount determined by a predetermined continuation correction function, and a correction selecting means for selecting whether or not to correct the allowable path error value in the teaching data. If the allowable path error value is 0, the continuous time change waveform of each axis of the robot is output to the outside, and if it is not 0, the correction continuous time obtained by adding the continuous correction amount to the continuous time change waveform. The trajectory generating apparatus for a robot according to claim 6, wherein the changed waveform is output to the outside.

With this, it is possible to realize a robot trajectory generation device which generates a curved trajectory that smoothly passes inside the second teaching point. Further, in the present invention according to claim 8, the continuation correction amount determined by the rate of change between the first time change waveform and the second time change at the second teaching point and a predetermined continuation correction function is used as the second teaching point. Path error calculating means for calculating the maximum path error between the second trajectory and the motion trajectory of the tip of the robot by the corrected continuous time change waveform, and the maximum path error is not less than the allowable path error of the teaching point data. And a continuous time resetting means for resetting the continuous start time and the continuous end time, and the maximum path error exceeds the allowable path error due to the continuous time reset by the continuous time resetting means. The trajectory generating apparatus for a robot according to any one of claims 6 to 7, wherein the trajectory generating apparatus does not satisfy the requirement.

Thus, it is possible to realize a robot trajectory generation device capable of resetting the continuation time so that the maximum path error does not exceed the allowable path error. Further, in the present invention according to claim 9, the continuation time resetting means sets the continuation start time point and the continuation end time point and then performs the arithmetic processing for obtaining the continuation time change waveform and the corrected continuation time change waveform. The robot trajectory generation device according to claim 8, wherein the continuation time is reset.

With this, it is possible to realize a robot trajectory generation device capable of resetting the continuation time while avoiding unnecessary processing.

[0019]

BEST MODE FOR CARRYING OUT THE INVENTION The present invention according to claim 1 is a teaching point for providing a reference of a trajectory along which a tip of a robot moves,
A movement command that specifies a trajectory shape based on the teaching point, a movement speed that specifies the speed at which the tip moves on the trajectory, and an allowance that specifies the maximum value of the error between the teaching point and the trajectory. A continuous first teaching point, a second teaching point on a teaching trajectory designated by teaching point data including a path error,
And for the third teaching point, the time of the angle of each axis of the robot corresponding to the movement from the first teaching point to the third teaching point is calculated by the interpolation calculation according to the trajectory shape and the moving speed designated by the teaching point data. The change waveform and the moving time are obtained, and based on the moving time, the passing points called the continuation start time point before the second teaching point and the continuation end time point after the second teaching point are separated from the second teaching point by the same time. The time point is set, and the first time change waveform of each axis of the robot corresponding to the movement from the first teaching point to the second teaching point is touched at the second teaching point and linearly changes until the end point of the continuation. The first extension time change waveform given by a line segment and the second time change waveform of each axis of the robot corresponding to the movement from the second teaching point to the third teaching point are in contact with the second teaching point at the second teaching point. Linearize until the start of conversion Obtains a second extended time change waveform given by line segments of, wherein the first period changing waveform second
An internal division point of the extension time change waveform and an internal division point of the second time change waveform and the first extension time change waveform are obtained by an internal division ratio by a predetermined continuation function, and the continuation function is the By using an S-shaped smooth function that gives an internal division ratio of 0 at the start of the continuation and an internal division ratio of 1 at the end of the continuation, the internal ratio is changed from the start of the continuation to the end of the continuation. A trajectory generation method for a robot, wherein a movement trajectory of the robot according to an angle of each robot axis given by an equidistant point is a continuous trajectory that smoothly passes through the second teaching point and is smoothly connected to the teaching trajectory. Further, the present invention according to claim 2 internally divides the continuation correction amount based on the slope change rate of the first time-varying waveform and the second time-varying waveform at the second teaching point and a predetermined continuation correction function. Correction to each axis of the robot by adding points The value of the continuous time change waveform is used, and the robot trajectory corresponding to the corrected continuous time change waveform of each axis is a corrected continuous trajectory that goes around the second teaching point and is smoothly connected to the taught trajectory. The robot trajectory generation method according to claim 1, and the present invention according to claim 3 is such that the continuation start time point and the continuation end time point are 0, and the maximum value is obtained when the second teaching point is passed. The trajectory generation of the robot according to claim 2, wherein the continuous correction function is defined as a continuous correction function, and a value weighted with a gradient change rate of the first time-varying waveform and the second time-varying waveform at the second teaching point is used as the continuous correction amount. The present invention according to claim 4 obtains a continuation correction amount at an angle of each robot axis corresponding to a set continuation disclosure time point and a continuation end time point set near a second teaching point, Based on the continuation correction amount The maximum path error between the trajectory and the corrected continuous trajectory is calculated, and when the maximum path error is equal to or more than the allowable route error given by the teaching point data, the continuation start time point and the continuation end time point are reset, The robot trajectory generation method according to any one of claims 2 to 3, wherein the maximum path error is not more than the allowable path error, and the present invention according to claim 5 is a continuous disclosure. The trajectory generation method for a robot according to claim 4, wherein after setting the time point and the continuation end time point, the continuation time is reset before the arithmetic processing for obtaining the continuation time change waveform and the corrected continuation time waveform. The present invention according to claim 6 provides a teaching point that gives a reference of a trajectory along which the tip of the robot moves, a movement command that specifies a trajectory shape based on the teaching point, and the tip on the trajectory. Department moved A continuous first teaching point on the teaching trajectory designated by the teaching point data having a moving velocity designating a velocity to be controlled and an allowable path error designating the maximum value of the error between the teaching point and the trajectory. , The second teaching point and the third teaching point are interpolated according to the trajectory shape and the moving speed designated by the teaching point data, and the third teaching point to the third teaching point are calculated.
Interpolation calculating means for calculating a time change waveform of the angle of each axis of the robot corresponding to the movement to the teaching point, and the second teaching point before the second teaching point based on the movement time from the first teaching point to the third teaching point. Continuation time setting means for setting a continuation start time point and a passing time point, which will be referred to as a continuation end time point later, at the time points equidistant from the second teaching point, and from the first teaching point to the second teaching point. A first extension time change waveform given by a line segment that is in contact with the first time change waveform of each axis of the robot corresponding to the movement of the robot at a second teaching point and linearly changes until the end of the continuation, and the second teaching The second extension time given by a line segment that touches the second time change waveform of each axis of the robot corresponding to the movement from the point to the third teaching point at the second teaching point and linearly changes until the continuation start time point Find the change waveform and
An internal division point of the first time variation waveform and the second extension time variation waveform and an internal division point of the second time variation waveform and the first extension time variation waveform are internally divided by a predetermined continuation function. A time-varying waveform continuation means for obtaining a continuous time-varying waveform of the angle of each axis of the robot obtained by the ratio, the continuation function having an internal division ratio of 0 at the start of the continuation and at the end of the continuation. By using an S-shaped smooth function that gives the internal division ratio as 1, the movement trajectory of the robot according to the angle of each axis of the robot given by the internal division point from the continuation start time to the continuation end time, A trajectory generating device for a robot, which is a continuous trajectory that smoothly passes through the second teaching point and is smoothly connected to the teaching trajectory, and
The present invention according to claim 7 is a continuation correction means for obtaining a continuation correction amount determined by a change rate of a first time change waveform and a second time change at a second teaching point and a predetermined continuation correction function. And a correction selecting means for selecting whether or not to correct the allowable path error value in the teaching data. The correction selecting means, when the allowable path error value is 0, the continuation time of each axis of the robot. 7. The trajectory generation of the robot according to claim 6, wherein the change waveform is output to the outside, and if it is not 0, the corrected continuous time change waveform obtained by adding the continuation correction amount to the continuous time change waveform is output to the outside. The present invention according to claim 8 provides a continuation correction amount determined by a change rate of a first time change waveform and a second time change waveform at a second teaching point and a predetermined continuation correction function. Obtained around the second teaching point Path error calculating means for calculating the maximum path error between the motion trajectory of the tip of the robot hand and the second teaching point by the corrected continuous time change waveform, and when the maximum path error is equal to or larger than the allowable path error of the teaching point data. Equipped with a continuation time resetting means for resetting the continuation start time and the continuation end time so that the maximum path error does not exceed the allowable path error due to the continuation time reset by the continuation time resetting means. The robot trajectory generation device according to any one of claims 6 to 7, and
In the present invention described in (1), the continuation time resetting means sets the continuation time before setting the continuation start time point and the continuation end time point and then calculating the continuation time change waveform and the corrected continuation time change waveform. The robot trajectory generation device according to claim 8, wherein the trajectory is reset.

Embodiments of the present invention will be described below. (Embodiment 1) An embodiment of a robot trajectory generation method according to the present invention will be described below with reference to the drawings. FIG. 1 (a) is a plan view showing the movement trajectory of the tip of the robot when moving between three teaching points by the trajectory generation method of the present invention, and FIG. 1 (b) is a time change of the position of the joint axis J1 of the robot. FIG. Note that, in FIG. 1A, the trajectory shape of the teaching trajectory x is shown by a straight line, but the trajectory generation method of the present invention also works effectively with other trajectory shapes.

In FIG. 1B, the horizontal axis t represents the elapsed time after the tip of the hand of the robot passes the teaching point P1, and the vertical axis θ represents the position of the joint axis J1 at the time t. In the description, the angle is used, but another expression method may be used. When the tip of the hand of the robot passes through the teaching point P1 in FIG. 1A, the angle of the joint axis J1 becomes θ1, and the teaching points P2 and θ2 and P3 and θ3 correspond to each other. Now, let us say that the time-varying waveform of the joint axis when moving from the teaching point P1 to the teaching point P2 is q1, and the time-varying waveform when moving from P2 to P3 is q2. Although the time-varying waveform is drawn as a straight line in FIG. 1B for simplicity, it is generally a curved line. Ts is the start point of continuation and Tp is the end point of continuation, and this setting will be described later. qn1
Is a line segment that has the same slope as the slope a1 of θ2 of the time change waveform q1 and that changes linearly from the teaching point passage time point Tp to the continuation end time point Te, that is, the line segment that is tangent to the time change waveform q1 at the teaching point passage time point Tp. Qn2 has the same slope as the slope a2 of θ2 of the time-varying waveform q2, and is a line segment that linearly changes from the continuation start time Ts to the teaching point passing time Tp, that is, the waveform q2 and the teaching point passing time Tp are in contact with each other. It is a line segment and is hereinafter referred to as an extended time change waveform.

The trajectory generation process will be described in detail below with reference to the drawings. Now, the moving time from the teaching point P1 to the teaching point p2 is set to T1, and the teaching point P2 to the teaching point P3.
When the moving time to is T2, the continuation start time Ts
And the continuation end time Te are determined by the following equations (1) and (2).

Ts = Tp-min (T1 / 2, T2 / 2) (1) Te = Tp + min (T1 / 2, T2 / 2) (2) where min (x , Y) is a function that selects the smaller value of x and y. From the above equations (1) and (2), Ts
And Te means that the continuation process is not performed for more than half of the shorter moving time when moving in the section before and after the teaching point P2.

Next, as shown in FIG. 1 (b), from the continuation start time Ts to the teaching point passage time Tp, there is a transition between the time variation waveform q1 and the extension time variation waveform qn2, and also the passage of the teaching point. From the time point Tp to the end time point Te of continuation, the internal division point is calculated by the following equations (3) and (4) between the time change waveform q2 and the extended time change waveform qn1, and the locus of the internal division point Is a new time-varying waveform, that is, a continuous time-varying waveform q. The internal division point is the continuation time Tc from the continuation start time Ts to the continuation end time Te, as shown in FIG.
As shown in (a), a smooth continuation function α (t) whose value continuously changes in an s-shape from 0 to 1 is used, and the elapsed time t is from the continuation start time Ts to the teaching point passage time. Equation (3) is used up to Tp, and the elapsed time t is the teaching point passing time T
From p to the end of the continuation, the above q is calculated by the equation (4).
Since the continuous time change waveform q is a time function, q
If (t), then q (t) = α (t) · qn2 (t) + (1-α (t)) · q1 (t) ·· (3) q (t) = α (t) · q2 (t) + (1−α (t)) · qn1 (t) ·· (4). The smoothness means that there is no discontinuity in the rate of change of the gradient of the continuation function, that is, the second derivative function.

The above processing is called continuous processing. By performing this continuous processing independently for each axis, the trajectory of the tip of the hand of the robot is the teaching point P2 as shown by x0 shown in FIG. 1 (a).
Is a continuous curve, which means that the polygonal line is continuous into a continuous curve.

Next, as shown in FIG. 1A, in order to correct and convert the continuous trajectory x0 into the inner trajectory xi, the continuation correction amount dq to be corrected for each joint axis is obtained. . The continuation correction amount dq is the inner constant k determined by the equation (5) from the slope a1 of the time-varying waveform q1 at θ2 and the slope a2 of the time-varying waveform q2, and the function shown in FIG. That is, the value is determined by the continuation correction function β (t), which has the maximum value at the teaching point passage time Tp during the continuation time Tc, as in the equation (6).

K = 2 (a2-a1) ... (5) dq = k.beta. (T) ... (6) This continuous correction amount dq is expressed by the equations (3) and (4). By adding to the obtained continuous time change waveform q, the corrected continuous time change waveform, that is, the corrected continuous time change waveform qi is obtained as shown in FIG. The inner circumference constant k is determined by the slope of the time-varying waveform q1 and the time-varying waveform q2.
The value is proportional to the difference between and, and can be said to indicate the rate of change of the slope. Therefore, the larger the change in the slope of the time-varying waveform q1 and the time-varying waveform q2 of each axis of the robot is, the larger the continuation correction amount is. It means that the amount is corrected to be large. The above process is referred to as a continuous correction process. By performing this continuous correction processing independently for each axis, the trajectory of the tip of the hand of the robot is
As indicated by xi in (a), the curve is a continuous curve passing inside the teaching point P2.

Next, the maximum path error Dr from the teaching point P2 of the inner circumference trajectory obtained by the above-mentioned continuation correction processing is obtained. Now, let us say that the vector representing the position of the teaching point P2 is x2. Continuation correction amount d of each axis at the time point Tp of passing the teaching point
The value of q is calculated in advance before the continuation process of each axis (this can be done independently of the continuation process as can be seen from the above formulas (5) and (6)) The angle of each axis of the robot after the correction processing is qqi (the target robot is 6
If it has an axis, qqi is a six-dimensional vector), and the passing vector position xi2 of the tip of the hand of the robot when the continuation correction process is performed is obtained by the equation (7).

Xi2 = F (qqi) (7) Here, the function F (v) is a function peculiar to the robot that obtains the tip position from the joint angle vector v of the robot. This corrected vector position xi2 and the teaching point P2 vector position x2
From the above, the maximum path error Dr can be calculated by the equation (8).

Dr = ‖xi2−x2‖ (8) Here, ‖z‖ represents a function for calculating the norm of the vector z.

Next, when the maximum path error Dr becomes equal to or larger than the allowable path error Da, the continuation time Tc is reset so as to narrow the continuation range. That is, when the maximum path error Dr becomes a value equal to or larger than the allowable path error Da of the teaching point data, the continuation time Tc is reset according to the equation (9).

Tc = [Tc · Dr / Da] (9) where [x] is a function that rounds down the decimal point if the real number x is 1 or more and is 1 if it is less than 1. . From the reset continuation time Tc, a new continuation start time Ts and a new continuation end time Te are calculated by using equations (10) and (1), respectively.
1) and the formula.

Ts = Tp−Tc / 2 (10) Te = Tp + Tc / 2 (11) Here, the reset continuation start time Ts and continuation end time Te are used. , Continuous processing and continuous correction processing are performed. This continuation time resetting process is performed at the continuation start time Ts.
Immediately after setting and the continuation end time point Te, by executing using only the continuation correction amount near the second teaching point, useless continuation processing and continuation correction processing when the path error is large are avoided. be able to. When the robot tip position can be calculated by the vector position of the angle of each axis of the robot as described above, it is needless to say that the continuation correction amount can be calculated only in the vicinity of the second teaching point, and the calculation processing amount can be reduced. Yes.

As described above, according to the present embodiment, by performing the continuation processing independently for each axis, the trajectory of the tip of the hand of the robot is converted into a continuous curve through which the teaching point passes, and
By performing the continuity correction processing independently for each axis, the trajectory of the tip of the robot's hand can be converted into a continuous curve that passes inside the teaching point, eliminating the need for acceleration / deceleration processing at the teaching point and takt time. Can be shortened. Also,
By resetting the continuation start time point and continuation end time point from the allowable path error before performing these continuation processing and continuation correction processing, the maximum path error between the trajectory of the tip of the robot hand and the position of the teaching point Can be a value equal to or less than the allowable path error.

(Embodiment 2) An embodiment of a robot trajectory generation apparatus according to the present invention will be described below with reference to the drawings. FIG. 3 is a block diagram showing the configuration of a trajectory generation device that executes the robot trajectory generation method of the present invention. The trajectory generation device of the present embodiment executes the trajectory continuation processing and the continuity correction processing according to the trajectory generation method described in the first embodiment. In the figure, 1
Is an interpolation calculation means, 2 is a continuation time setting means, 3 is a path error calculation means, 4 is a continuation time resetting means, 5 is a time-varying waveform continuation means, 6 is a continuation correction means, and 7 is a correction selection means. Is.

The mutual relationship and operation of the above-mentioned components will be described. Interpolation calculation means 1 inputs teaching point data, performs interpolation calculation according to a designated trajectory shape and moving speed, and changes time-varying waveforms q1 and q2 of angles of each axis of the robot with continuation time setting means 2 and time-varying waveform. It is output to the serializing means 5.
The continuation time setting means 2 inputs the time-varying waveforms q1 and q2 of the angles of the robot axes from the interpolation calculation means 1, and based on them, the continuation start time point Ts and the continuation end time point Te described in (1) above. Equation (2) and the waveform of the angle of each axis of the robot with time are output to the path error calculating means 3. The path error calculation means 3 inputs the teaching point data, the time change waveforms q1 and q2 of the angles of the robot axes, the continuation start time Ts, and the continuation end time Te, and the continuation correction process is performed. Position x of the robot's hand tip
i2 is obtained by the equation (7), the maximum path error Dr with respect to the teaching point P2 is calculated by the equation (8), and is output to the continuation time resetting means 4 together with the continuation start time Ts and the continuation end time Te.

If the maximum path error Dr is equal to or larger than the allowable path error Da of the teaching point data, the continuation time resetting means 4
The continuation start time point Ts and the continuation end time point Te are expressed by equation (9),
It is reset by the equations (10) and (11) and output to the time-varying waveform continuation means 5. Time-varying waveform continuation means 5
Is the time-varying waveform q1, q2 of the angle of each robot axis, the continuation start time point Ts, and the continuation end time point Te from the interpolation calculation means 1, and calculates the continuous time-varying waveform q of the angle of each robot axis. Then, the continuous correction means 6 and the correction selection means 7
And the continuation start time Ts and the continuation end time Te are output to the continuation correction means 6. The continuation correction means 6 obtains the continuation correction amount dq for the continuous time change waveform q of the angle of each axis of the robot, and the correction selection means 7
Output to The continuous correction amount for calculating the maximum path error Dr may be calculated only in the vicinity of the second teaching point,
As a result, the amount of calculation processing can be reduced.

The correction selecting means 7 is a teaching point data, a continuous time change waveform q of the angle of each axis of the robot, and a continuous correction amount d.
Input q and the allowable path error value Da of the teaching point data is 0
Then, the continuous time-varying waveform q of the angle of each axis of the robot obtained by the time-varying waveform continuous means 5 is output to the outside as it is, and if the allowable path error value Da of the teaching point data is not 0, it is continuous. The continuation correction amount dq obtained by the correction means 6 is added to the continuation time change waveform q and corrected to output qi as a corrected continuation time change waveform. By operating the robot according to the time-varying waveform of the angle of each axis of the robot that is output to the outside, the operation of the robot tip portion is smooth and the teaching point is the same as in the method of generating the trajectory of the robot described in the first embodiment. Therefore, the acceleration / deceleration processing in can be eliminated.

Needless to say, each constituent element of this embodiment and its operation can be realized by a program operation of a microcomputer.

[0040]

As is apparent from the above description, according to the present invention, the teaching point is smoothly passed through the trajectory of the tip of the hand of the robot by independently performing the continuation process on each axis of the robot. It can be a continuous curve. Further, by performing the continuation correction processing independently for each axis, it is possible to form a smooth continuous curve in which the trajectory of the tip of the hand of the robot passes inside the teaching point. Further, by the continuation process and the continuation correction process, the acceleration / deceleration process at the teaching point is unnecessary, and the tact time can be shortened. In addition, by resetting the continuation start time and continuation end time based on the allowable path error before performing the continuation process and the continuation correction process, the trajectory of the tip of the robot hand and the position of the teaching point The process of making the maximum path error smaller than the allowable path error can be performed while avoiding unnecessary processing.

[Brief description of the drawings]

FIG. 1 is a plan view showing a trajectory of a tip of a robot hand and a characteristic diagram showing a temporal change of an angular position of each joint axis of a robot in a method for generating a trajectory of a robot of the present invention.

FIG. 2 is a characteristic diagram showing a continuation function and a continuation correction function according to the present invention.

FIG. 3 is a block diagram showing the configuration of a robot trajectory generation device of the present invention.

[Explanation of symbols]

 1 interpolation calculation means 2 continuation time setting means 3 path error calculation means 4 continuation time resetting means 5 time-varying waveform continuation means 6 continuation correction means 7 correction selection means P1 teaching point (first teaching point) P2 teaching point (Second teaching point) P3 Teaching point (Third teaching point) θ1, θ2, θ3 Joint angle at teaching point q1 Time change waveform (First time change waveform) q2 Time change waveform (Second time change waveform) q Continuous Time change waveform qi Corrected continuous time change waveform qn1 Extension time change waveform (first extension time change waveform) qn2 Extension time change waveform (second extension time change waveform) x Teaching trajectory x0 Continuous trajectory xi Corrected continuous trajectory Dr Max Route error

────────────────────────────────────────────────── ───

[Procedure amendment]

[Submission date] February 7, 1996

[Procedure amendment 1]

[Document name to be amended] Statement

[Correction target item name] 0021

[Correction method] Change

[Correction contents]

In FIG. 1B, the horizontal axis t represents the elapsed time after the tip of the hand of the robot passes the teaching point P1, and the vertical axis θ represents the position of the joint axis J1 at the time t. In the description, the angle is used, but another expression method may be used. When the tip of the hand of the robot passes through the teaching point P1 in FIG. 1A, the angle of the joint axis J1 becomes θ1, and the teaching points P2 and θ2 and P3 and θ3 correspond to each other. Now, let us say that the time-varying waveform of the joint axis when moving from the teaching point P1 to the teaching point P2 is q1, and the time-varying waveform when moving from P2 to P3 is q2. Although the time-varying waveform is drawn as a straight line in FIG. 1B for simplicity, it is generally a curved line. Ts is the start point of continuation and Te is the end point of continuation, and this setting will be described later. qn1
Is a line segment that has the same slope as the slope a1 of θ2 of the time change waveform q1 and that changes linearly from the teaching point passage time point Tp to the continuation end time point Te, that is, the line segment that is tangent to the time change waveform q1 at the teaching point passage time point Tp. Qn2 has the same slope as the slope a2 of θ2 of the time-varying waveform q2, and is a line segment that linearly changes from the continuation start time Ts to the teaching point passing time Tp, that is, the waveform q2 and the teaching point passing time Tp are in contact with each other. It is a line segment and is hereinafter referred to as an extended time change waveform.

[Procedure amendment 2]

[Document name to be amended] Statement

[Correction target item name] 0022

[Correction method] Change

[Correction contents]

The trajectory generation process will be described in detail below with reference to the drawings. Now, from teaching point P1 to teaching point
The movement time to P2 is T1, and the teaching point P2 to teaching point P3
When the moving time to is T2, the continuation start time Ts
And the continuation end time Te are determined by the following equations (1) and (2).

[Procedure 3]

[Document name to be amended] Statement

[Name of item to be corrected] 0025

[Correction method] Change

[Correction contents]

The above processing is called continuous processing. By performing the continuous process in each axis independently, as X ₀ trajectory of the hand tip of the robot illustrated in FIG. 1 (a), the teaching point P2
Is a continuous curve, which means that the polygonal line is continuous into a continuous curve.

[Procedure amendment 4]

[Document name to be amended] Statement

[Correction target item name] 0026

[Correction method] Change

[Correction contents]

Next, as shown in FIG. 1A, in order to correct and convert the continuous trajectory X ₀ into the inner trajectory xi, the continuation correction amount dq to be corrected for each joint axis is calculated. Ask. The continuation correction amount dq is the inner constant k determined by the equation (5) from the slope a1 of the time-varying waveform q1 at θ2 and the slope a2 of the time-varying waveform q2, and the function shown in FIG. That is, the value is determined by the continuation correction function β (t), which has the maximum value at the teaching point passage time Tp during the continuation time Tc, as in the equation (6).

[Procedure Amendment 5]

[Document name to be amended] Statement

[Correction target item name] 0032

[Correction method] Change

[Correction contents]

Tc = Tc · Da / Dr ... (9) Here, [x] is a function that rounds down the decimal point when the real number x is 1 or more and is 1 when it is less than 1. From the reset continuation time Tc, a new continuation start time Ts and a new continuation end time Te are calculated by using equations (10) and (1 ), respectively.
It is determined by the formula 1) .

Claims (9)

[Claims] 1. A teaching point that gives a reference of a trajectory along which a tip of a robot moves, a movement command that specifies a trajectory shape based on the teaching point, and a movement that specifies a speed at which the tip moves on the trajectory. A continuous first teaching point and a second teaching point on the teaching trajectory designated by the teaching point data having a velocity and an allowable path error designating the maximum value of the error between the teaching point and the trajectory; And for the third teaching point, the time of the angle of each axis of the robot corresponding to the movement from the first teaching point to the third teaching point is calculated by the interpolation calculation according to the trajectory shape and the moving speed designated by the teaching point data. The change waveform and the moving time are obtained, and based on the moving time, the passing points called the continuation start time point before the second teaching point and the continuation end time point after the second teaching point are separated from the second teaching point by the same time. Set to time Then, with a line segment that touches the first time-varying waveform of each axis of the robot corresponding to the movement from the first teaching point to the second teaching point at the second teaching point and linearly changes until the end of the continuation. The given first extension time change waveform and the second time change waveform of each axis of the robot corresponding to the movement from the second teach point to the third teach point are contacted at the second teach point at the continuation start time point. A second extension time change waveform given by a line segment that linearly changes up to, and an internal division point between the first time change waveform and the second extension time change waveform,
An internal division point of the second time-varying waveform and the first extension time-varying waveform is obtained by an internal division ratio by a predetermined continuation function,
The continuation function is an S-shaped smooth function that gives an internal division ratio of 0 at the start of the continuation and an internal division ratio of 1 at the end of the continuation, so that the continuation function starts from the continuation start time. The robot's movement trajectory according to the angle of each axis of the robot given by the internal division point until the end of conversion is a continuous trajectory that smoothly passes through the second teaching point and is smoothly connected to the teaching trajectory. Orbit generation method. 2. A robot each by adding a continuation correction amount based on a slope change rate of the first time-varying waveform and the second time-varying waveform at a second teaching point and a predetermined continuation correction function to the internally dividing point. The value of the corrected continuous time change waveform of the axis is set, and the robot trajectory corresponding to the corrected continuous time change waveform of each axis is
The trajectory generation method for a robot according to claim 1, wherein a corrected continuous trajectory is provided that goes inside the taught point and is smoothly connected to the taught trajectory. 3. A smoothing function having 0 at the continuation start time point and the continuation end time point and having a maximum value at the time point when the second teaching point has passed is defined as a continuation correction function, and a first time change waveform at the second teaching point is set. The robot trajectory generation method according to claim 2, wherein a value weighted by a gradient change rate with respect to the second temporal change waveform is used as the continuation correction amount. 4. A continuation correction amount at an angle of each axis of the robot corresponding to a set continuation disclosure time point and a continuation end time point is determined near a second teaching point, and a teaching trajectory is obtained based on the continuation correction amount. The maximum path error with the corrected continuous trajectory is calculated, and when the maximum path error is equal to or more than the allowable path error given by the teaching point data, the continuation start time point and the continuation end time point are reset to set the maximum path error. The robot trajectory generation method according to claim 2 or 3, wherein an error is set not to exceed the allowable path error. 5. The continuation time is reset before the arithmetic processing for obtaining the continuation time change waveform and the corrected continuation time waveform after setting the continuation disclosure time point and the continuation end time point. The trajectory generation method of the described robot. 6. A teaching point which gives a reference of a trajectory along which the tip of the robot moves, a movement command which designates a trajectory shape based on the teaching point, and a movement which designates a speed at which the tip moves on the trajectory. A continuous first teaching point and a second teaching point on the teaching trajectory designated by the teaching point data having a velocity and an allowable path error designating the maximum value of the error between the teaching point and the trajectory; And for the third teaching point, the time of the angle of each axis of the robot corresponding to the movement from the first teaching point to the third teaching point is calculated by the interpolation calculation according to the trajectory shape and the moving speed designated by the teaching point data. Interpolation calculation means for calculating a change waveform, and the third teaching point from the first teaching point
Based on the moving time to the teaching point, a continuation time for setting a passing point called a continuation starting time point before the second teaching point and a continuation ending time point later after the second teaching point, respectively at a time point that is equidistant from the second teaching point. Setting means and the second teaching point from the first teaching point
A first extension time change waveform given by a line segment that contacts the first time change waveform of each axis of the robot corresponding to the movement to the teach point at a second teach point and linearly changes until the end point of the continuation; A line segment that is in contact with the second time-varying waveform of each axis of the robot corresponding to the movement from the second teaching point to the third teaching point at the second teaching point and linearly changes until the continuation start time is given. 2 extension time change waveform is obtained, and an internal division point between the first time change waveform and the second extension time change waveform,
A time-varying waveform sequence in which an internal division point of the second time-varying waveform and the first extension time-varying waveform is obtained by an internal division ratio by a predetermined continuation function to be a continuous time-varying waveform of the angle of each axis of the robot. And a continuation function, and the continuation function is an S-shaped smooth function that gives an internal division ratio of 0 at the start of the continuation and an internal division ratio of 1 at the end of the continuation. From the start point of conversion to the end point of continuation, the movement trajectory of the robot according to the angle of each axis of the robot given by the internal division point smoothly passes through the second teaching point and is smoothly connected to the teaching trajectory. A trajectory generation device for a robot that uses a modified trajectory. 7. A continuation correction means for obtaining a continuation correction amount determined by a rate of change between a first time change waveform and a second time change at a second teaching point and a predetermined continuation correction function,
Correction selection means for selecting whether or not to correct the allowable path error value in the teaching data, wherein the correction selection means
When the allowable path error value is 0, the continuous time change waveform of each axis of the robot is output to the outside, and when it is not 0, the corrected continuous time change obtained by adding the continuous correction amount to the continuous time change waveform. The robot trajectory generation device according to claim 6, wherein the waveform is output to the outside. 8. A continuation correction amount determined near the second teaching point by obtaining a continuation correction amount determined by a change rate of the first time change waveform and the second time change at the second teaching point and a predetermined continuation correction function. Path error calculating means for calculating the maximum path error between the second trajectory and the motion trajectory of the tip of the robot by the time-varying waveform, and continuous when the maximum path error is equal to or larger than the allowable path error of the teaching point data. A request for resetting a start time point and a continuation end time point, and a continuous time resetting means for preventing the maximum path error from exceeding the allowable path error due to the continuous time reset by the continuous time resetting means. The trajectory generation device for a robot according to claim 6 or 7. 9. The continuation time resetting means resets the continuation time before setting the continuation start time point and the continuation end time point and then calculating the continuation time change waveform and the corrected continuation time change waveform. The robot trajectory generation device according to claim 8.